Benchmark
WildClawBench · Multimodal (26 tasks)
Scores are percentages. Higher is better.
Benchmark sourceScores from BenchBoard docs.
- Models and systems measured
- 38
- Scores recorded
- 38
- Sources
- 1
- IQ weight
- 1.21
Score distribution
- 41.1%
- Mean
- 40.7%
- Median
- 15.9%
- Std dev
- 23.8%
- 10th percentile
- 63%
- 90th percentile
- 7.3%
- Min
This benchmark has a weight of 1.21 for chat models, measured over 17 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model or system | Score |
|---|---|---|
| 1 | Qwen3.8 Omni Flash多模态任务子集 · Claude Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 71% |
| 2 | GPT-5.6 SolOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 69.3% |
| 3 | Claude Fable 5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 68.6% |
| 4 | Claude Opus 4.8OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 65.3% |
| 5 | GPT 5.5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 63% |
| 6 | Qwen3.8 MaxOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 59.4% |
| 7 | Gemini 3.8 Flash多模态任务子集 · Claude Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 58.9% |
| 8 | Claude Opus 4.7OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 58.5% |
| 9 | Grok 4.5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 55.3% |
| 10 | Muse Spark 1.1OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 53.7% |
| 11 | GLM 5.2OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 50.9% |
| 12 | Claude Opus 4.6OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 47.7% |
| 13 | Muse Glimmer 30BOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 46.6% |
| 14 | Kimi K3OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 43.8% |
| 15 | Gemini 3.1 Pro PreviewOpenClaw · multimodal-input split · 26 tasks · low-effort· Reported by WildClawBench / InternLM· via BenchBoard docs | 43.5% |
| 16 | Hy3 (Deep Infra)OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 43.1% |
| 17 | Kimi K2.7 CodeOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 41.7% |
| 18 | GLM 5.1OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 40.7% |
| 19 | GPT 5.4OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 40.2% |
| 20 | Qwen 3.8 27BOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 39.1% |
| 21 | Qwen3.6 27BOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 38.2% |
| 22 | MiMo V2.5 ProOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 36.2% |
| 23 | Gemma 4 31B ITOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 34.7% |
| 24 | Qwen3.5 Omni Plus多模态任务子集 · Claude Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 34.5% |
| 25 | MiMo V2 FlashOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 34.3% |
| 26 | DeepSeek V4 Pro 0813OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 33.6% |
| 27 | GLM 5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 30.7% |
| 28 | MiMo V2 ProOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 29.7% |
| 29 | DeepSeek V3.2OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 26.1% |
| 30 | Kimi K2.5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 24% |
| 31 | Qwen3.5 397B A17BOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 23.8% |
| 32 | MiniMax M2.7OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 19.2% |
| 33 | MiniMax M2.5OpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 16.3% |
| 34 | Step 3.5 FlashOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 12.6% |
Not ranked
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| GLM 5 TurboOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 24.5% |
| Grok 4.20 BetaOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 7.3% |
| Intern-S2 Preview 397BOpenClaw · multimodal-input split · 26 tasks· Reported by WildClawBench / InternLM· via BenchBoard docs | 34.2% |
| Seed-2.0-Lite多模态任务子集 · Claude Code· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 41.9% |